Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

Dataset: 80% C4 (Web Text) and 20% Python Code

Neuron 395 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.0649. Min Range: -0.0649

Max Act: 0.0591. Min Act: -0.0002

Data Index: 990709 (C4 (Web Text))

Max Activating Token Index: 682

Text #1

Max Range: 0.0649. Min Range: -0.0649

Max Act: 0.0649. Min Act: -0.0002

Data Index: 50718 (C4 (Web Text))

Max Activating Token Index: 89

Text #2

Max Range: 0.0649. Min Range: -0.0649

Max Act: 0.0642. Min Act: -0.0002

Data Index: 48424 (C4 (Web Text))

Max Activating Token Index: 996

Text #3

Max Range: 0.0649. Min Range: -0.0649

Max Act: 0.0632. Min Act: -0.0002

Data Index: 39247 (C4 (Web Text))

Max Activating Token Index: 344

Text #4

Max Range: 0.0649. Min Range: -0.0649

Max Act: 0.0627. Min Act: -0.0002

Data Index: 758351 (C4 (Web Text))

Max Activating Token Index: 330

Text #5

Max Range: 0.0649. Min Range: -0.0649

Max Act: 0.0552. Min Act: -0.0002

Data Index: 1087985 (C4 (Web Text))

Max Activating Token Index: 788

Text #6

Max Range: 0.0649. Min Range: -0.0649

Max Act: 0.0644. Min Act: -0.0002

Data Index: 874860 (C4 (Web Text))

Max Activating Token Index: 571

Text #7

Max Range: 0.0649. Min Range: -0.0649

Max Act: 0.0594. Min Act: -0.0002

Data Index: 1320701 (C4 (Web Text))

Max Activating Token Index: 204

Text #8

Max Range: 0.0649. Min Range: -0.0649

Max Act: 0.0572. Min Act: -0.0002

Data Index: 200713 (C4 (Web Text))

Max Activating Token Index: 134

Text #9

Max Range: 0.0649. Min Range: -0.0649

Max Act: 0.0563. Min Act: -0.0002

Data Index: 879628 (C4 (Web Text))

Max Activating Token Index: 557

Text #10

Max Range: 0.0649. Min Range: -0.0649

Max Act: 0.0535. Min Act: -0.0002

Data Index: 1084239 (C4 (Web Text))

Max Activating Token Index: 247

Text #11

Max Range: 0.0649. Min Range: -0.0649

Max Act: 0.0566. Min Act: -0.0002

Data Index: 1138946 (C4 (Web Text))

Max Activating Token Index: 519

Text #12

Max Range: 0.0649. Min Range: -0.0649

Max Act: 0.0589. Min Act: -0.0002

Data Index: 823678 (C4 (Web Text))

Max Activating Token Index: 248

Text #13

Max Range: 0.0649. Min Range: -0.0649

Max Act: 0.0552. Min Act: -0.0002

Data Index: 1096851 (C4 (Web Text))

Max Activating Token Index: 162

Text #14

Max Range: 0.0649. Min Range: -0.0649

Max Act: 0.0548. Min Act: -0.0002

Data Index: 1309756 (C4 (Web Text))

Max Activating Token Index: 505

Text #15

Max Range: 0.0649. Min Range: -0.0649

Max Act: 0.0571. Min Act: -0.0002

Data Index: 1099857 (C4 (Web Text))

Max Activating Token Index: 396

Text #16

Max Range: 0.0649. Min Range: -0.0649

Max Act: 0.0569. Min Act: -0.0002

Data Index: 284221 (C4 (Web Text))

Max Activating Token Index: 562

Text #17

Max Range: 0.0649. Min Range: -0.0649

Max Act: 0.0530. Min Act: -0.0002

Data Index: 1210964 (C4 (Web Text))

Max Activating Token Index: 333

Text #18

Max Range: 0.0649. Min Range: -0.0649

Max Act: 0.0546. Min Act: -0.0002

Data Index: 1035245 (C4 (Web Text))

Max Activating Token Index: 269

Text #19

Max Range: 0.0649. Min Range: -0.0649

Max Act: 0.0559. Min Act: -0.0002

Data Index: 348166 (C4 (Web Text))

Max Activating Token Index: 64